DSXchange

Each job will generate N * (M + 1) + 1 processes, where N is the number of nodes and M is the number of operators (approximately the same as the number of stages). You say when you start monitoring that there were already 300 osh processes, and your job caused this to jump to 856. So clearly your jo...

If you can not come up with a specification in words, please get a business analyst to help you to do so. We could guess at your requirement, but could end up wasting time if the guess is wrong. My guess is that you want to overwrite all but the last record in each group with YYYY in the non-key col...

Don't apply StringToDate() within the IsValid() function - the IsValid() function expects a string argument (not a Date argument).

What other warnings/errors are logged?

Is it likely that the DataStage server/network/database server is overloaded causing timeouts to trigger?

Whatever environment variable names you like, though I counsel meaningful names. Use these to supply values to the properties in the Teradata Connector stage.

Why is it in an aborted stage? Which event in the job log caused it to abort? Too many warnings generated? You can adjust that at run time. When you add a reject link to your Oracle Enterprise stage (and set the Output Reject Record property to True), the sqlcode column is automatically added to the...

The Stored Procedure stage calls stored procedures, not cursors. You would need to create a stored procedure that DataStage can call (that is, one that handles a result set) and possibly call this from a database stage rather than a Stored Procedure stage. Or you can simply use the SELECT statement ...

Note that this limit is imposed by your mail client, not by DataStage.

Can you write the file to a shared drive and send a link to it?

DataSet1 does not have an input, so its partitioning will be governed by whatever it was set to in the job that created the Data Set. Partitioning on the input link to the Funnel stage from DataSet1 should be Same. Partitioning on the input link to the Funnel stage from Oracle should be the same as ...

You don't (have range lookup in version 7.5). Craig's not working on DataStage at the moment. Join is only ever equijoin, so you won't achieve your requirement that way. Would you consider using a server job, where you can use user-defined SQL in the database stage that services the reference input ...

Does your Transformer stage have a reject-handling link?

Not all of your source data are in valid YYYYMMDD format.

Then use "%yyyy%mm%dd" (without hyphens) as the format string in StringToDate.

You don't need any particular format to send to Oracle - that it is a Date data type means that it will be handled correctly.

How long does Ora1 take to sort 25million+ rows if DataStage is not involved? Is the column on which you're sorting indexed using a B-tree index (which would help a lot)?

I imagine that U could specify how to do it, but you don't need Perl (which is what I assume you mean by "pl"). Please note that U does not login to DSXchange all that often. The second person personal pronoun in English is spelled "you". I expect the final word in your sentence ...